Return-Path: Received: from corot.bc.edu (corot.bc.edu [136.167.2.209]) by monet.bc.edu (8.8.7/8.8.7) with ESMTP id NAA68348 for ; Thu, 30 Sep 1999 13:40:04 -0400 From: maiser@efs.mq.edu.au Received: (from root@localhost) by corot.bc.edu (8.8.7/8.8.7) with X.500 id NAA59866 for baum@mail1.bc.edu; Thu, 30 Sep 1999 13:40:04 -0400 Received: from baldrick.ocs.mq.edu.au (baldrick.ocs.mq.edu.au [137.111.1.12]) by corot.bc.edu (8.8.7/8.8.7) with ESMTP id NAA214936 for ; Thu, 30 Sep 1999 13:39:58 -0400 Received: from efs1.efs.mq.edu.au (efs1.efs.mq.edu.au [137.111.64.8]) by baldrick.ocs.mq.edu.au (8.9.2/8.9.2) with ESMTP id DAA12828 for ; Fri, 1 Oct 1999 03:39:53 +1000 (EST) Received: from EFS1/SpoolDir by efs1.efs.mq.edu.au (Mercury 1.40); 1 Oct 99 03:42:33 GMT+1000 Received: from SpoolDir by EFS1 (Mercury 1.40); 1 Oct 99 03:42:31 GMT+1000 To: baum@bc.edu Date: Fri, 1 Oct 99 3:42:31 GMT+1000 Subject: Re: Message-ID: <2B1458654E@efs1.efs.mq.edu.au> From: "Jian Yang" To: "RATS Discussion List" Subject: Re: stability in VECM Date: Tue, 31 Aug 1999 09:08:03 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: Novell GroupWise 5.5.2 (via Mercury MTS (Bindery) v1.40) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable See p.281 in "Long Run Economic Relationships". There is a convergence = condition. Hope it is useful to you. Jian Yang >>> Price SG 08/31/99 05:27AM >>> In single equation ECMs the stability condition on the loading is obviously pretty simple. But in VECM the condition for stability/convergence to the long-run equilibria are more complex. In particular, I imagine we might occasionally encounter models where the estimated loading(s) in an ECM equation in the VECM has the "wrong" ie positive sign (when the level corresponding to the lhs variable is normalised with a positive coefficient), but the system is still stable.=20= Does anyone know any papers that discuss this? ---------- End of message ---------- From: "Jongchurl Kim" To: "RATS Discussion List" Subject: notice Date: Sat, 4 Sep 1999 15:26:54 -0400 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 Content-Type: text/plain; Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Outlook Express 4.72.3110.1 (via Mercury MTS (Bindery) v1.40) Hello, I am in your mailing list. I would like to let you know my new email address. From kim.317@osu.edu to sonac2@hanmail.net Please send any new mails to the latter address. Thanks a lot. ---------- End of message ---------- From: Florian =?iso-8859-1?Q?H=F6ppner?= To: "RATS Discussion List" Subject: var.src Date: Tue, 07 Sep 1999 10:12:48 +0200 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Institute for International Economics, University of Bonn X-Mailer: Mozilla 4.5 [en] (Win95; I) (via Mercury MTS (Bindery) v1.40) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Dear RATS user, I would like to input my own P-Matrix for a structural decomposition of the residuals in the procedure var.src. I have choosen the option INPUT but I don=B4t know how and where to input my own P-Matrix. Does anybody know? Thanks a lot in advance, Florian Hoeppner ---------- End of message ---------- From: Victor Sula To: "RATS Discussion List" Subject: Re: stability in VECM Date: Wed, 8 Sep 1999 02:03:43 -0700 (PDT) Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mercury MTS (Bindery) v1.40 unsubscribe __________________________________________________ Do You Yahoo!? Bid and sell for free at http://auctions.yahoo.com ---------- End of message ---------- From: Petri Mäki-Fränti To: "RATS Discussion List" Subject: Re: var.src Date: Thu, 09 Sep 1999 16:19:05 +0300 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) (via Mercury MTS (Bindery) v1.40) Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable At 10:12 7.9.1999 +0200, you wrote: >Dear RATS user, > >I would like to input my own P-Matrix for a structural decomposition of >the residuals in the procedure var.src. I have choosen the option INPUT >but I don=B4t know how and where to input my own P-Matrix. Does anybody >know? >Thanks a lot in advance, > >Florian Hoeppner > I think, that the structuralization you mean, is denoted by B0 in the instructions of the var.src procedure: VAR start end B0 * # (list of the series in the VAR) * # (OPTIONAL list of exogenous series (w/ EXOG option)) You have to define the B0-matrix before you run the var.src -procedure in= the usual way, by declare- and input- commands. Petri M=E4ki-Fr=E4nti ---------- End of message ---------- From: kohns@iiw.uni-bonn.de To: "RATS Discussion List" Subject: unit-root tests with uradf.src and additional regressors Date: Fri, 10 Sep 1999 17:12:12 +0200 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.01d) (via Mercury MTS (Bindery) v1.40) Dear RATS users, I would like to perform unit root tests on a variety of series and I need to control for deterministic seasonal effects, a structural break (due to a change in the way the data were collected) and a time trend. Can this be done within the framework of URADF.SRC? My problem is the following: 1. I cannot include additional determinstic regressors directly in the regressions performed by URADF.SRC. 2. I tried to use the residuals from a previous regression of the series on the determenistic regressors. But then I believe that I should set the option DET=NONE, since a constant and a trend have already been included in the previous regression, and doing so, URADF will not give me the proper critical values, since I still have to look at the \tau_\tau test statistic (and not the \tau statistic for regressions without intercept or trend). 3. If I still set the option DET=TREND (even though the constant and the trend have already been removed), I get the proper critical values, but the coefficients on the constant and the trend are not equal to zero, which makes me doubt whether the results are still valid. Tanks a lot for your help, Stephan Kohns *************************************************** * Dipl.-Volkswirt Stephan Kohns * * Institute for International Economics * * University of Bonn * * Germany * * Tel.: 0049-228-739234 * * E-Mail: kohns@iiw.uni-bonn.de * *************************************************** ---------- End of message ---------- From: Katja Gerling To: "RATS Discussion List" Subject: SMPL in panel data Date: Tue, 14 Sep 1999 17:34:33 +0200 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: The Kiel Institute of World Economics X-Mailer: Mozilla 4.6 [en] (Win98; I) (via Mercury MTS (Bindery) v1.40) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear all, I have a question concerning the SMPL command when working with panel data: if I have a panel with, say, 10 members and annual data from 1990 to 1998 and I want to cut out the first year for all panel members, i.e. estimate a regression from 1991 to 1998, how do I set the SMPL command right? I tried several ways, but it does not seem to work out... Thanks, Katja ---------- End of message ---------- From: "Sergio" To: "RATS Discussion List" Subject: non linear restrictions Date: Tue, 14 Sep 1999 12:58:02 -0600 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 Content-Type: multipart/alternative; X-Mailer: Microsoft Outlook Express 4.72.3110.5 (via Mercury MTS (Bindery) v1.40) This is a multi-part message in MIME format. ------=_NextPart_000_0027_01BEFEB0.C7A91A00 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Rats users: In order to test non linear restrictions on the coefficients, generally = Wald type test is used. (Judge et al. Introd... 1988, Eq (12.3.59) ) I think that is not performed directly in RATS, because Handbook page = 6-4 refers only a set of Q "linear" restrictions. Is that right? So, Does anybody have some piece of code for that?. Thanks in advance, *************************************************** Sergio Zuniga szuniga@ucn.cl =20 Universidad Catolica del Norte =20 Coquimbo - Chile - Tel.: 56-51-327248 =20 *************************************************** ------=_NextPart_000_0027_01BEFEB0.C7A91A00 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Dear Rats users:
In order to = test non linear=20 restrictions on the coefficients, generally Wald type test is = used.
(Judge et al. Introd... 1988, Eq (12.3.59) = )
I think that is not performed directly in RATS, = because=20 Handbook page 6-4 refers only a set of Q "linear"=20 restrictions.
Is that right? So, Does = anybody have some=20 piece of code for that?.
Thanks in advance,
 
***************************************************
Sergio=20 Zuniga       szuniga@ucn.cl    =         =20
Universidad Catolica del Norte 
Coquimbo - Chile - Tel.:=20 56-51-327248          &= nbsp;           &n= bsp;=20
***************************************************
 
------=_NextPart_000_0027_01BEFEB0.C7A91A00-- ---------- End of message ---------- From: Woo Kai-Yin To: "RATS Discussion List" Subject: RE:FM-GMM Date: Wed, 15 Sep 1999 01:46:31 +0800 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: Mozilla 3.01C-IMS (Win95; I) (via Mercury MTS (Bindery) v1.40) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear RATS users, Might I request for codes to implement Kitamura's FM-GMM estimation and Wald test statistics, and Quintos' FM-GMM reduced rank estimation and wald test statistics, please? I feel grateful if you could give me a hand. regardes, KY ---------- End of message ---------- From: "Estima" To: "RATS Discussion List" Subject: Re: non linear restrictions Date: Tue, 14 Sep 1999 13:40:33 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Estima MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.11) (via Mercury MTS (Bindery) v1.40) > Dear Rats users: > In order to test non linear restrictions on the coefficients, generally = > Wald type test is used. > (Judge et al. Introd... 1988, Eq (12.3.59) ) > I think that is not performed directly in RATS, because Handbook page = > 6-4 refers only a set of Q "linear" restrictions. > Is that right? So, Does anybody have some piece of code for that?. > Thanks in advance, Many non-linear restrictions can be linearized, and thus handled using RESTRICT (or MRESTRICT). For example, suppose you want to impose the restriction: Beta1*Beta2 = some value X To linearize this, take partial differentials: Beta1*Beta2 approx. = Beta1Hat*Beta2Hat + $ || Beta2Hat, Beta1Hat ||*|| Beta1 - Beta1Hat | Beta2 - Beta2hat || Where Beta1 and Beta2 are the "true" values, and the "Hat" terms are the estimated parameters. Beta1*Beta2 approx. = Beta2Hat*Beta1 + $ Beta1Hat*Beta2 - Beta1Hat*Beta2Hat Beta1*Beta2 approx. = Beta2Hat*Beta1 + $ Beta1Hat*Beta2 - Beta1Hat*Beta2Hat So the restriction is: X + Beta1Hat*Beta2Hat = $ Beta2Hat*Beta1 + Beta1Hat*Beta2 Let's assume that you are using the names "B1" and "B2" as the variables for Beta1 and Beta2, respectively (i.e. these are the names you use on NONLIN and in your formulas). Further, assume that B1 is coefficient #1 and B2 is #2 (i.e. that these are the first two coefficients listed on the NONLIN). Once you've done your estimation, you can implement this as: restrict 1 # 1 2 # b2 b1 (b1*b2+x) Note that the estimated value of B2 is thus the weight on the term for B1, and estimated B1 is the weight for the B2 term. Sincerely, Tom Maycock Estima -- ------------------------------------------------------------ | Estima | Sales: (800) 822-8038 | | P.O. Box 1818 | (847) 864-8772 | | Evanston, IL 60204-1818 | Support: (847) 864-1910 | | USA | Fax: (847) 864-6221 | | http://www.estima.com | estima@estima.com | ------------------------------------------------------------ ---------- End of message ---------- From: "Guido Travaglini" To: "RATS Discussion List" Subject: Re: SMPL in panel data Date: Tue, 14 Sep 1999 18:23:50 PDT Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Mime-Version: 1.0 Content-Type: text/plain; format=flowed X-Mailer: Mercury MTS (Bindery) v1.40 Dear Katja, suppose you have i=1...n individuals and t=1...T observations (balanced panel) and want to estimate series X for j=1...mFrom: Katja Gerling >Reply-To: "RATS Discussion List" >To: "RATS Discussion List" >Subject: SMPL in panel data >Date: Tue, 14 Sep 1999 17:34:33 +0200 > >Dear all, >I have a question concerning the SMPL command when working with panel >data: if I have a panel with, say, 10 members and annual data from 1990 >to 1998 and I want to cut out the first year for all panel members, i.e. >estimate a regression from 1991 to 1998, how do I set the SMPL command >right? I tried several ways, but it does not seem to work out... >Thanks, Katja > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ---------- End of message ---------- From: "Guido Travaglini" To: "RATS Discussion List" Subject: Complex matrix Date: Wed, 15 Sep 1999 02:50:25 PDT Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Mime-Version: 1.0 Content-Type: text/plain; format=flowed X-Mailer: Mercury MTS (Bindery) v1.40 Anybody can tell me how to compute the determinant and the inverse of a matrix with complex numbers? I have tried to declare first the complex scalar and matrix (e.g. dec rec[complex] ...) but it doesn't work. I've also tried to use, to no avail however, some of the complex number functions listed in the manual. Help! Thanx, Guido. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ---------- End of message ---------- From: "Estima" To: "RATS Discussion List" Subject: Re: SMPL in panel data Date: Wed, 15 Sep 1999 10:30:05 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Estima MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.11) (via Mercury MTS (Bindery) v1.40) > suppose you have i=1...n individuals and t=1...T observations (balanced > panel) and want to estimate series X for j=1...m something like this to cut your series: > do i = 1,n > do j = 1,m (or do j = m,T) > smpl i//j i//j > dis(...) X(i//j) > end do j > end do i Nope, I'm afraid that's not going to work. First, I'll try to explain why, then we'll look at ways that will work. The basic issue is that SMPL command when used in the form: SMPL start end can only specify a contiguous range of entries (i.e. no gaps), from the "start" entry to the "end" entry. Thus, because a panel series is stored as a continuous "stacked" variable, with one individual following another, there's no way to omit a set of observations from each individual--because that would involve gaps in the series. Further, you can only have one "SMPL" in effect at any given time. So, rather than build up a complex sample range, the code sample above is actually just resetting the SMPL each time through the loop. Fortunately, there are at least two easy ways to specify a sample range with gaps. The most general method is to use SET to define the appropriate dummy variable, and then use the SMPL option (available on most estimation instructions and many others dealing with series) to control the range. Note that this works with cross section and time series data as well as panel data. For panel data, though, you've got the %period and %indiv functions which provide easy ways to set up smples excluding certain time periods and/or individuals. Suppse you want to drop the first period from each individual in a regression. Just do: set mysample = %period(t)>1 The defines a series with 1's everywhere except for the first observation for each individual, which will be set to zero. Note that :%period(t)<>1" or "%period(t)>=2" would also work. linreg(smpl=mysample) y # x z You can even use the SMPL command to make this the "default" sample range, by doing: smpl(sample=mysample) Note, however, that this setting only affects those instructions that also have an SMPL option--it's just the equivalent of using the SMPL option on each instruction individual. See Chap 5, p. 2 and Chap 12, p. 13, and the description of SMPL in Chap 14, for more info. If you have your panel data stored in a RATS format file (and particularly if you have dates assigned to the series), you can control which entries from each individual are read into RATS. For example, suppose your series contain monthly data from 1990:1 through 1998:12 for each individual, and you want to drop 1990:1 from each individual for your analysis. You could either read in all the data and set a dummy as described above, or you can just set your CALENDAR to start in 1990:2 (be sure to set the PANELOBS option accordingly--in this case to 107 observations rather than 108): cal(panel=107) 1990 2 12 allocate 5//1998:12 open data panel.rat data(for=rats,verbose) Note that this only works if the RATS file has dates associated with the series you are reading. If your RATS file contains undated panel data, you can't drop observations from the beginning of the range, but you can drop from them the end (by specifying a smaller value for PANELOBS). Hope this is helpful. Tom Maycock Estima -- ------------------------------------------------------------ | Estima | Sales: (800) 822-8038 | | P.O. Box 1818 | (847) 864-8772 | | Evanston, IL 60204-1818 | Support: (847) 864-1910 | | USA | Fax: (847) 864-6221 | | http://www.estima.com | estima@estima.com | ------------------------------------------------------------ ---------- End of message ---------- From: "Estima" To: "RATS Discussion List" Subject: Re: Complex matrix Date: Wed, 15 Sep 1999 15:16:03 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Estima MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.11) (via Mercury MTS (Bindery) v1.40) > Anybody can tell me how to compute the determinant and the inverse > of a matrix with complex numbers? I have tried to declare first the > complex scalar and matrix (e.g. dec rec[complex] ...) but it doesn't > work. I've also tried to use, to no avail however, some of the complex > number functions listed in the manual. Help! Thanx, Guido. Guido: You can certainly define complex arrays. For example, here's a crude example: dec rect[complex] rc(2,2) ewise rc(i,j) = %cmplx(1.0,0.0) write rc However, the built-in INV() and DET() functions are not capable of handling complex arguments, so any such computation would have to be written by hand (and I have to admit I have no idea how difficult that would be, never having tried it). Sincerely, Tom Maycock Estima -- ------------------------------------------------------------ | Estima | Sales: (800) 822-8038 | | P.O. Box 1818 | (847) 864-8772 | | Evanston, IL 60204-1818 | Support: (847) 864-1910 | | USA | Fax: (847) 864-6221 | | http://www.estima.com | estima@estima.com | ------------------------------------------------------------ ---------- End of message ---------- From: Rob Trevor To: "RATS Discussion List" Subject: Re: Complex matrix Date: Thu, 16 Sep 1999 09:15:06 +1000 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Mailer: Mercury MTS (Bindery) v1.40 > > Anybody can tell me how to compute the determinant and the inverse >> of a matrix with complex numbers? I have tried to declare first the >> complex scalar and matrix (e.g. dec rec[complex] ...) but it doesn't >> work. I've also tried to use, to no avail however, some of the complex > > number functions listed in the manual. Help! Thanx, Guido. >... > >You can certainly define complex arrays. For example, here's a >... >However, the built-in INV() and DET() functions are not capable of >handling complex arguments, so any such computation would have >to be written by hand (and I have to admit I have no idea how >difficult that would be, never having tried it). It is not so much difficult as tedious! If you don't know how to do the row/column operations yourself, check out any book like Numerical Recipes for should tell you how to do it. The tedious part is translating that into RATS code. Rob Trevor ---------- End of message ---------- From: "J.E. Sturm" To: "RATS Discussion List" Subject: Random-effects probit model Date: Thu, 16 Sep 1999 10:17:35 +0100 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Economie RUG MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.01d) (via Mercury MTS (Bindery) v1.40) Hi, I have a probit model and a panel data set. I would like to estimate the model using random effects. Does anyone have any experience with this? If yes, then please send me an e-mail. Thanks in advance. Jan-Egbert Sturm University of Groningen Faculty of Economics Department of General Economics P.O.Box 800 9700 AV Groningen The Netherlands Phone: +31 50 3634538 Fax: +31 50 3637337 E-mail: J.E.Sturm@eco.rug.nl Url: http://www.eco.rug.nl/medewerk/sturm/ ---------- End of message ---------- From: Galindo Andrade Arturo To: "RATS Discussion List" Subject: Impulse Response Function in CATS Date: Mon, 20 Sep 1999 17:43:27 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) (via Mercury MTS (Bindery) v1.40) Content-Type: text/plain; Content-Transfer-Encoding: quoted-printable > Dear Rats and Cats users, >=20 > I would like to know if someone can help me in the construction of = impulse > response functions in a VEC model that I have estimated in CATS. Does = any > one know of a procedure to do so? >=20 > In advance thanks a lot, >=20 > Arturo Galindo=20 Banco de la Rep=FAblica Bogot=E1, Colombia ---------- End of message ---------- From: Juan Toro To: "RATS Discussion List" Subject: Re: Impulse Response Function in CATS Date: Tue, 21 Sep 1999 15:04:22 +0200 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: Mozilla 4.5 [en] (WinNT; I) (via Mercury MTS (Bindery) v1.40) MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Dear Arturo: MALCOLM for RATS does it. Chers Galindo Andrade Arturo wrote: > > Dear Rats and Cats users, > > > > I would like to know if someone can help me in the construction of im= pulse > > response functions in a VEC model that I have estimated in CATS. Does= any > > one know of a procedure to do so? > > > > In advance thanks a lot, > > > > Arturo Galindo > Banco de la Rep=FAblica > Bogot=E1, Colombia -- Juan Toro PhD Student Economics Department European University Institute Badia Fiesolana I-50016 San Domenico de Fiesole Italy Phone Work :00/39/55/599086 Phone Home :00/39/55/4685271 ---------- End of message ---------- From: "VIJAYA SARATHI N." To: "RATS Discussion List" Subject: Date: Thu, 23 Sep 1999 02:38:18 GMT+0530 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: IIM, Ahmedabad X-mailer: Pegasus Mail v3.31 (via Mercury MTS (Bindery) v1.40) hello i'm trying to use the procedure crosspec.src. eventhough i source the code, i get the error message cp18 "crosspec is not the name of a procedure. did you forget to source" src file is in the right directory, has the right name and extension. Could you help me out with the possible remedies please. regards, Sarathi ---------- End of message ---------- From: "Ravindra A. Yatawara" To: "RATS Discussion List" Subject: Re: Impulse Response Function using var.src Date: Thu, 23 Sep 1999 17:10:15 -0400 (EDT) Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Mailer: Mercury MTS (Bindery) v1.40 Hi List, A couple of quick questions re. a blanchard quah decomposition. 1) i used the var.src to get bootstrapped IRFs. i get graphs that are titled Effects of a shock to inflation i) inflation picture here ii) growth picture does the above mean i) impact of demand shocks on inflation ii) impact of demand shocks on growth? sorry if this seems trivial - just want to make sure. 2) how do i get the demand shocks and supply shocks? i tried going to the var function and chhosing residual autocorrelation -there i got VAR residuals for growth ( a line graph) and a bar graph with autocorrelation info. does this VAR resdiduals for growth respond to "supply shocks"?? if not - do you know how i can get them? and the demands shocks as well. thanks very much ravi ____________________ Ravi A. Yatawara Department of Economics 1012 International Affairs Building Columbia University New York, NY 10027 Tel: 212-666-5445 Fax: 212-854-8059 email: ray1@columbia.edu ---------- End of message ---------- From: mike mccracken To: "RATS Discussion List" Subject: roots of polynomial Date: Thu, 23 Sep 1999 17:28:13 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) (via Mercury MTS (Bindery) v1.40) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" List members: Using RATS, is there a simple way to determine the roots of the autoregressive and moving average polynomial(s) implied by the estimated coefficients from (say) the Boxjenk procedure? I know E-views spits it out automatically. If this is simple to do it would be nice if the Boxjenk procedure had it as an option. Thanks in advance. Mike *************************************************************** Michael W. McCracken Assistant Professor of Economics ph: (225) 388-3782 Department of Economics fax: (225) 388-3807 Louisiana State University 2107 CEBA Bldg. Baton Rouge, LA 70803-6306 *************************************************************** ---------- End of message ---------- From: "Estima" To: "RATS Discussion List" Subject: Warning: fake messages re. Y2K compliance Date: Fri, 24 Sep 1999 12:42:49 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: Organization: Estima MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT X-mailer: Pegasus Mail for Win32 (v3.11) (via Mercury MTS (Bindery) v1.40) Dear RATS User's: We just heard from a RATS user who received a bogus message claiming to be from Estima, asking the recipient to go to a web site to run a Y2K compliance check. This message was most definitely _not_ sent by Estima. At this point, we don't yet know whether this was an isolated practical joke targeted at this particular person, or some wider spam attempt. In any case, if you do receive such a message please let us know. I would strongly recommend that you not visit any web site mentioned or respond to the e-mail. I've reprinted the text of the original message below, along with the indicated "From:" address. Sincerely, Tom Maycock Estima ************************************************************ From: assistance@ihls.com To: Subject: Quick Y2K Compliance Check Date sent: Thu, 23 Sep 1999 11:58:40 -0700 We recently send you RATS for Windows 4.2. To ensure compatability with your system, please run this Y2K compliancy test for your computer. To do so, visit this web site address: http://www.ihls.com/y2k.htm?BS=BS990923_42920.246 19219 Your Estima team, Tom -- ------------------------------------------------------------ | Estima | Sales: (800) 822-8038 | | P.O. Box 1818 | (847) 864-8772 | | Evanston, IL 60204-1818 | Support: (847) 864-1910 | | USA | Fax: (847) 864-6221 | | http://www.estima.com | estima@estima.com | ------------------------------------------------------------ ---------- End of message ---------- From: "Jian Yang" To: "RATS Discussion List" Subject: Re: sufficient cointegration Date: Wed, 29 Sep 1999 15:29:13 -0500 Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: X-Mailer: Novell GroupWise 5.5.2 (via Mercury MTS (Bindery) v1.40) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Dear RATS users, I have a question about Toda and Phillips (1993). If the sufficeint = cointegration requirement cannot be met, the noncausality test even based = on ECM fails. But, does the relevant weak exogeneity test, which is a part = of the noncausality test, also break down? I would like to have more = information on this topic. Thank you for your help. Jian Yang Ph.D. candidate Texas A&M University=20 ---------- End of message ---------- From: SATHOMPSON@CCTR.UMKC.EDU To: "RATS Discussion List" Subject: Re: Impulse Response Function using var.src Date: Wed, 29 Sep 1999 19:06:58 -0500 (CDT) Errors-to: Reply-to: "RATS Discussion List" Sender: Maiser@efs1.efs.mq.edu.au X-listname: MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII X-Mailer: Mercury MTS (Bindery) v1.40 From: IN%"RATS-L@efs.mq.edu.au" "RATS Discussion List" 23-SEP-1999 16:24:02.97 To: IN%"RATS-L@efs.mq.edu.au" "RATS Discussion List" CC: Subj: RE: Impulse Response Function using var.src Return-path: Received: from baldrick.ocs.mq.edu.au (baldrick.ocs.mq.edu.au [137.111.1.12]) by CCTR.UMKC.EDU (PMDF V5.2-31 #37767) with ESMTP id <01JGBE39AZOAFZLPI7@CCTR.UMKC.EDU> for SATHOMPSON@CCTR.UMKC.EDU (ORCPT rfc822;SATHOMPSON@CCTR.UMKC.EDU); Thu, 23 Sep 1999 16:23:03 CDT Received: from efs1.efs.mq.edu.au (efs1.efs.mq.edu.au [137.111.64.8]) by baldrick.ocs.mq.edu.au (8.9.2/8.9.2) with ESMTP id HAA08611; Fri, 24 Sep 1999 07:12:28 +1000 (EST) Received: from EFS1/SpoolDir by efs1.efs.mq.edu.au (Mercury 1.40); Fri, 24 Sep 1999 07:15:21 +0000 Received: from SpoolDir by EFS1 (Mercury 1.40); Fri, 24 Sep 1999 07:13:05 +0000 Date: Thu, 23 Sep 1999 17:10:15 -0400 (EDT) From: "Ravindra A. Yatawara" Subject: Re: Impulse Response Function using var.src Sender: Maiser@efs1.efs.mq.edu.au To: RATS Discussion List Errors-to: rats-l-owner@efs.mq.edu.au Reply-to: RATS Discussion List Message-id: <16788FC6593@efs1.efs.mq.edu.au> MIME-version: 1.0 X-Mailer: Mercury MTS (Bindery) v1.40 Content-type: TEXT/PLAIN; charset=US-ASCII X-Listname: Original-recipient: rfc822;SATHOMPSON@CCTR.UMKC.EDU Hi List, A couple of quick questions re. a blanchard quah decomposition. 1) i used the var.src to get bootstrapped IRFs. i get graphs that are titled Effects of a shock to inflation i) inflation picture here ii) growth picture does the above mean i) impact of demand shocks on inflation ii) impact of demand shocks on growth? sorry if this seems trivial - just want to make sure. 2) how do i get the demand shocks and supply shocks? i tried going to the var function and chhosing residual autocorrelation -there i got VAR residuals for growth ( a line graph) and a bar graph with autocorrelation info. does this VAR resdiduals for growth respond to "supply shocks"?? if not - do you know how i can get them? and the demands shocks as well. thanks very much ravi ____________________ Ravi A. Yatawara Department of Economics 1012 International Affairs Building Columbia University New York, NY 10027 Tel: 212-666-5445 Fax: 212-854-8059 email: ray1@columbia.edu ---------- End of message ----------